visual-edit

Pass

Audited by Gen Agent Trust Hub on Jul 13, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Fetches and executes the @agent-native/core package via npx. This tool is part of the vendor's ecosystem and is required for the live-editing functionality.
  • [COMMAND_EXECUTION]: Utilizes shell commands including npx and curl to manage the local bridge process and verify its manifest.
  • [PROMPT_INJECTION]: Potential for indirect prompt injection through the processing of local application route manifests.
  • Ingestion points: Application manifest data (routes and capabilities) retrieved from the local dev server.
  • Boundary markers: None explicitly implemented when passing manifest data to subsequent actions.
  • Capability inventory: Includes the ability to write to the local file system via the write-local-file action and perform network operations.
  • Sanitization: The skill explicitly states that human write consent remains mandatory, providing a critical check against automated malicious actions.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 13, 2026, 04:51 AM
Security Audit — agent-trust-hub — visual-edit